Wileybook

#!/bin/bash                              
# Call this script like
#   Wileybook http://onlinelibrary.wiley.com/book/10.1002/9780470316887 or (mandatory for some books)
#   Wileybook http://onlinelibrary.wiley.com/book/10.1002/0471705195
# where the final digits are the ISBN numbers (possibly with 978 - the publisher missing)

# Parse the arguments
if [ ! $# = 1 ]; then
	echo "Usage: $(basename $0) URL-to-Wiley-content"
	echo "Example: $(basename $0) http://onlinelibrary.wiley.com/book/10.1002/9780470316887"
	exit
fi

# First (and only) command line argument is the URL of the online resource
URL=$1                                                                     
ISBN=`echo $URL | grep -o '[0-9]\{10,\}'`
if [ ! $ISBN"xxx" == "xxx" ]; then
	echo "Looking up $URL (ISBN $ISBN)"
else
	echo "$(basename $0): Failed to parse ISBN."
	exit
fi

# Remember the current directory
pushd .

# Make a temporary directory 
echo "Making directory /tmp/$ISBN"
rm -rf /tmp/$ISBN
mkdir /tmp/$ISBN

# Get the source
cd /tmp/$ISBN
echo "$(basename $0): Getting the HTML source"
wget --user-agent="Mozilla/5.0" $URL -O source

# Parse the source to get the session ID
export SID=`grep -o 's=[a-f0-9]\{40\}' source`
echo "We have the session id $SID"

# Parse the source to retrieve an ordered list of download URLs
# grep -o '"/doi/10\.1002/$ISBN/pdf"' source > list
grep -o '/doi[a-zA-Z0-9./]*/pdf[[:alnum:]?=&;]*' source > list

rm -f logfile

echo "$(basename $0): Getting the individual chapter files"
for i in $(cat list); do 
	wget --user-agent="Mozilla/5.0" "http://onlinelibrary.wiley.com$i" -O source2
	j=$(grep -m 1 -o 'http://onlinelibrary.wiley.com/store/[a-zA-Z0-9.:/]*.pdf[[:alnum:]?=&;]*' source2)
	wget --user-agent="Mozilla/5.0" "$j" -a logfile
done

# Create a list of files to concatenate
grep -e "^Saving to:.*" logfile | grep --only-matching -e "[[:alnum:]-]*\.pdf[[:alnum:].?=&;]*" | uniq > download_list

# qpdf is confused by the name of the downloaded files.
# Therefore, we rename them.
j=0;
rm -f concat_list;
for i in $(cat download_list); do
	mv "$i" $((++j)).pdf;
	echo $j.pdf >> concat_list
done

BOOKTITLE=$ISBN

# Concatenate the pdf files to a book
echo "$(basename $0): Concatenating files"
qpdf --empty $BOOKTITLE.pdf --pages $(cat concat_list) --

# Return to the current directory
popd

# Inform the user
echo "The book has been compiled and now resides in /tmp/$ISBN/$BOOKTITLE.pdf"